The present invention relates to an automatic focusing camera for recording a motion image and a still image through automatic focusing and to an automatic focusing method used for such a camera.
A conventional digital camera capable of recording a motion image and a still image performs DCT (Discrete Cosine Transform) based on an image signal outputted from an image sensing device and performs automatic focusing such that a specified spatial frequency component is enhanced. More specifically, the DCT is performed on image signals for a plurality of frames, while shifting a focal position by driving a lens, and the lens is moved to a position at which, e.g., RF components are maximized (see, e.g., Japanese Laid-Open Patent Publication No. HEI 6-245129.
When the number of pixels composing one frame is increased, the time required for the reading of image signals from the image sensing device and the time required for the DCT are increased so that a time interval between the sensing of an image and the recording of a focused image tends to be longer. In view of the tendency, automatic focusing is performed based on a thinned image signal for an image from which pixels in, e.g., vertical directions have been thinned out and the foregoing thinned image signal is recorded for the recording of a motion image, while image signals for all the pixels are read out of the image sensing device in the state in which automatic focusing has been performed for the recording of a still image, by considering that the resolution of the motion image need not necessarily be the same as that of a still image.
When the pixels are thinned out as described above, however, the quality of the motion image is degraded since image information is basically missing. To prevent this, an improvement in the quality of a motion image may be considered by causing the image sensing device to output one mixed image signal obtained by mixing respective image signals for each specified number of pixels such that it is recorded. In this case, however, automatic focusing when a motion image is recorded is also performed based on the mixed image signal. In the automatic focusing based on the mixed image signal, however, it becomes difficult to record a still image which is focused with high precision when the recording of the still image is performed during the recording of a motion image. That is, since the mixed image signal includes image signals for all the pixels, while the mixing of the image signals lowers the RF components, high-precision automatic focusing suited to the recording of a still image cannot be performed.
Thus, the conventional automatic focusing camera encounters the problem that it is difficult to perform the recording of a still image which is automatically focused with high precision, while allowing the recording of a high-quality motion image.